Seeking an optimal approach for Computer-aided Diagnosis of Pulmonary Embolism
Pulmonary Embolism (PE) represents a thrombus ("blood clot"), usually originating from a lower extremity vein, that travels to the blood vessels in the lung, causing vascular obstruction and in some patients death. This disorder is commonly diagnosed using Computed Tomography Pulmonary Ang...
Saved in:
Published in | Medical image analysis Vol. 91; p. 102988 |
---|---|
Main Authors | , , , , |
Format | Journal Article |
Language | English |
Published |
Netherlands
01.01.2024
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Pulmonary Embolism (PE) represents a thrombus ("blood clot"), usually originating from a lower extremity vein, that travels to the blood vessels in the lung, causing vascular obstruction and in some patients death. This disorder is commonly diagnosed using Computed Tomography Pulmonary Angiography (CTPA). Deep learning holds great promise for the Computer-aided Diagnosis (CAD) of PE. However, numerous deep learning methods, such as Convolutional Neural Networks (CNN) and Transformer-based models, exist for a given task, causing great confusion regarding the development of CAD systems for PE. To address this confusion, we present a comprehensive analysis of competing deep learning methods applicable to PE diagnosis based on four datasets. First, we use the RSNA PE dataset, which includes (weak) slice-level and exam-level labels, for PE classification and diagnosis, respectively. At the slice level, we compare CNNs with the Vision Transformer (ViT) and the Swin Transformer. We also investigate the impact of self-supervised versus (fully) supervised ImageNet pre-training, and transfer learning over training models from scratch. Additionally, at the exam level, we compare sequence model learning with our proposed transformer-based architecture, Embedding-based ViT (E-ViT). For the second and third datasets, we utilize the CAD-PE Challenge Dataset and Ferdowsi University of Mashad's PE Dataset, where we convert (strong) clot-level masks into slice-level annotations to evaluate the optimal CNN model for slice-level PE classification. Finally, we use our in-house PE-CAD dataset, which contains (strong) clot-level masks. Here, we investigate the impact of our vessel-oriented image representations and self-supervised pre-training on PE false positive reduction at the clot level across image dimensions (2D, 2.5D, and 3D). Our experiments show that (1) transfer learning boosts performance despite differences between photographic images and CTPA scans; (2) self-supervised pre-training can surpass (fully) supervised pre-training; (3) transformer-based models demonstrate comparable performance but slower convergence compared with CNNs for slice-level PE classification; (4) model trained on the RSNA PE dataset demonstrates promising performance when tested on unseen datasets for slice-level PE classification; (5) our E-ViT framework excels in handling variable numbers of slices and outperforms sequence model learning for exam-level diagnosis; and (6) vessel-oriented image representation and self-supervised pre-training both enhance performance for PE false positive reduction across image dimensions. Our optimal approach surpasses state-of-the-art results on the RSNA PE dataset, enhancing AUC by 0.62% (slice-level) and 2.22% (exam-level). On our in-house PE-CAD dataset, 3D vessel-oriented images improve performance from 80.07% to 91.35%, a remarkable 11% gain. Codes are available at GitHub.com/JLiangLab/CAD_PE. |
---|---|
AbstractList | Pulmonary Embolism (PE) represents a thrombus ("blood clot"), usually originating from a lower extremity vein, that travels to the blood vessels in the lung, causing vascular obstruction and in some patients death. This disorder is commonly diagnosed using Computed Tomography Pulmonary Angiography (CTPA). Deep learning holds great promise for the Computer-aided Diagnosis (CAD) of PE. However, numerous deep learning methods, such as Convolutional Neural Networks (CNN) and Transformer-based models, exist for a given task, causing great confusion regarding the development of CAD systems for PE. To address this confusion, we present a comprehensive analysis of competing deep learning methods applicable to PE diagnosis based on four datasets. First, we use the RSNA PE dataset, which includes (weak) slice-level and exam-level labels, for PE classification and diagnosis, respectively. At the slice level, we compare CNNs with the Vision Transformer (ViT) and the Swin Transformer. We also investigate the impact of self-supervised versus (fully) supervised ImageNet pre-training, and transfer learning over training models from scratch. Additionally, at the exam level, we compare sequence model learning with our proposed transformer-based architecture, Embedding-based ViT (E-ViT). For the second and third datasets, we utilize the CAD-PE Challenge Dataset and Ferdowsi University of Mashad's PE Dataset, where we convert (strong) clot-level masks into slice-level annotations to evaluate the optimal CNN model for slice-level PE classification. Finally, we use our in-house PE-CAD dataset, which contains (strong) clot-level masks. Here, we investigate the impact of our vessel-oriented image representations and self-supervised pre-training on PE false positive reduction at the clot level across image dimensions (2D, 2.5D, and 3D). Our experiments show that (1) transfer learning boosts performance despite differences between photographic images and CTPA scans; (2) self-supervised pre-training can surpass (fully) supervised pre-training; (3) transformer-based models demonstrate comparable performance but slower convergence compared with CNNs for slice-level PE classification; (4) model trained on the RSNA PE dataset demonstrates promising performance when tested on unseen datasets for slice-level PE classification; (5) our E-ViT framework excels in handling variable numbers of slices and outperforms sequence model learning for exam-level diagnosis; and (6) vessel-oriented image representation and self-supervised pre-training both enhance performance for PE false positive reduction across image dimensions. Our optimal approach surpasses state-of-the-art results on the RSNA PE dataset, enhancing AUC by 0.62% (slice-level) and 2.22% (exam-level). On our in-house PE-CAD dataset, 3D vessel-oriented images improve performance from 80.07% to 91.35%, a remarkable 11% gain. Codes are available at GitHub.com/JLiangLab/CAD_PE. |
ArticleNumber | 102988 |
Author | Gotway, Michael B Liang, Jianming Zhou, Zongwei Islam, Nahid Ul Gehlot, Shiv |
Author_xml | – sequence: 1 givenname: Nahid Ul surname: Islam fullname: Islam, Nahid Ul organization: School of Computing and Augmented Intelligence, Arizona State University, Tempe, AZ 85281, USA – sequence: 2 givenname: Zongwei surname: Zhou fullname: Zhou, Zongwei organization: Department of Computer Science, Johns Hopkins University, Baltimore, MD 21218, USA – sequence: 3 givenname: Shiv surname: Gehlot fullname: Gehlot, Shiv organization: Biomedical Informants Program, Arizona State University, Phoenix, AZ 85054, USA – sequence: 4 givenname: Michael B surname: Gotway fullname: Gotway, Michael B organization: Department of Radiology, Mayo Clinic, Phoenix, AZ 85054, USA – sequence: 5 givenname: Jianming surname: Liang fullname: Liang, Jianming email: Jianming.Liang@asu.edu organization: Biomedical Informants Program, Arizona State University, Phoenix, AZ 85054, USA. Electronic address: Jianming.Liang@asu.edu |
BackLink | https://www.ncbi.nlm.nih.gov/pubmed/37924750$$D View this record in MEDLINE/PubMed |
BookMark | eNo9kMlOwzAQhi1URBd4AiTkI5cUL0lsH1Epi1QBEnC27MQuLkkc7OTQt29KS6WRZjTzz_ZNwajxjQHgGqM5Rji_28xrUzo1J4jQIUME52dggmmOE54SOjrFOBuDaYwbhBBLU3QBxpQJkrIMTcDrhzE_rllD1UDfdq5WFVRtG7wqvqH1AS583fadCYlypSnhg1PrxkcXobfwva9q36iwhcta-8rF-hKcW1VFc3X0M_D1uPxcPCert6eXxf0qKWgmuiQjgmmGrCkUEkxgorFgtMCDIZtaTblIDWIl40hZzYXOS8SszhWmoixERmfg9jB3uPS3N7GTtYuFqSrVGN9HSTjPKeFMpIOUHqRF8DEGY2UbhjfDVmIk9yDlRv6BlHuQ8gBy6Lo5Luj1UD31_JOjO7iMcc0 |
CitedBy_id | crossref_primary_10_1016_j_eswa_2023_123029 crossref_primary_10_1002_ima_23037 crossref_primary_10_3390_electronics13030476 crossref_primary_10_1007_s10140_024_02216_2 crossref_primary_10_2478_jce_2023_0019 crossref_primary_10_1111_eci_14139 |
Cites_doi | 10.1109/CVPR.2018.00745 10.1109/CVPR.2017.75 10.1109/CVPR46437.2021.01549 10.1109/CVPR42600.2020.00975 10.1148/radiology.186.3.8430170 10.1109/TMI.2016.2535302 10.1109/CVPR.2018.00393 10.1109/CVPR.2017.506 10.1109/CVPR.2017.243 10.1016/j.artmed.2017.03.008 10.1109/CVPR.2016.90 10.1118/1.3157102 10.1007/978-3-030-01264-9_9 10.1109/CVPR42600.2020.00674 10.1109/TMI.2015.2482920 10.1109/ICCV48922.2021.00951 10.1016/j.media.2019.101541 10.1109/ICCV48922.2021.00950 10.1016/j.media.2020.101840 10.1109/TBME.2010.2063702 10.1016/j.compbiomed.2007.10.001 10.1109/ACCESS.2019.2925210 10.1109/TMI.2021.3089292 10.1148/ryai.2021200254 10.1016/j.thromres.2012.11.027 10.1148/radiology.194.3.7862997 10.1109/CVPR.2017.195 10.1109/WACV56688.2023.00358 10.1109/CVPR46437.2021.00676 10.1038/sdata.2018.180 10.1016/j.media.2017.07.005 10.1109/CVPR.2017.634 10.1056/NEJMoa052367 10.1016/j.cmpb.2013.12.014 10.1378/chest.104.6.1685 10.1109/TMI.2016.2536809 |
ContentType | Journal Article |
Copyright | Copyright © 2023. Published by Elsevier B.V. |
Copyright_xml | – notice: Copyright © 2023. Published by Elsevier B.V. |
DBID | CGR CUY CVF ECM EIF NPM AAYXX CITATION 7X8 |
DOI | 10.1016/j.media.2023.102988 |
DatabaseName | Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed CrossRef MEDLINE - Academic |
DatabaseTitle | MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) CrossRef MEDLINE - Academic |
DatabaseTitleList | MEDLINE - Academic MEDLINE |
Database_xml | – sequence: 1 dbid: NPM name: PubMed url: https://proxy.k.utb.cz/login?url=http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: EIF name: MEDLINE url: https://proxy.k.utb.cz/login?url=https://www.webofscience.com/wos/medline/basic-search sourceTypes: Index Database |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Medicine Engineering |
EISSN | 1361-8423 |
EndPage | 102988 |
ExternalDocumentID | 10_1016_j_media_2023_102988 37924750 |
Genre | Journal Article |
GrantInformation_xml | – fundername: NHLBI NIH HHS grantid: R01 HL128785 |
GroupedDBID | --- --K --M .~1 0R~ 1B1 1~. 1~5 29M 4.4 457 4G. 53G 5GY 5VS 7-5 71M 8P~ AACTN AAEDT AAEDW AAIKJ AAKOC AALRI AAOAW AAQFI AAQXK AAXKI AAXUO AAYFN ABBOA ABBQC ABJNI ABMAC ABMZM ABXDB ACDAQ ACGFS ACIUM ACIWK ACNNM ACPRK ACRLP ACZNC ADBBV ADEZE ADJOM ADMUD ADTZH ADVLN AEBSH AECPX AEKER AENEX AFJKZ AFKWA AFRAH AFTJW AFXIZ AGHFR AGUBO AGYEJ AHJVU AHZHX AIALX AIEXJ AIKHN AITUG AJOXV AJRQY AKRWK ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ ANZVX AOUOD ASPBG AVWKF AXJTR AZFZN BJAXD BKOJK BLXMC BNPGV C45 CAG CGR COF CS3 CUY CVF DU5 EBS ECM EFJIC EIF EJD EO8 EO9 EP2 EP3 F5P FDB FEDTE FGOYB FIRID FNPLU FYGXN G-Q GBLVA GBOLZ HVGLF HX~ HZ~ IHE J1W JJJVA KOM M41 MO0 N9A NPM O-L O9- OAUVE OVD OZT P-8 P-9 P2P PC. Q38 R2- RIG ROL RPZ SDF SDG SDP SEL SES SEW SPC SPCBC SSH SST SSV SSZ T5K TEORI UHS ~G- AAYXX CITATION 7X8 |
ID | FETCH-LOGICAL-c359t-5297b70feca097912b1973c13c10f4fb3894e07d780afb89b6d07fb6a139dc953 |
ISSN | 1361-8415 |
IngestDate | Sat Aug 17 05:22:34 EDT 2024 Thu Sep 26 19:04:49 EDT 2024 Sun Oct 13 10:15:28 EDT 2024 |
IsPeerReviewed | true |
IsScholarly | true |
Keywords | Pulmonary embolism Swin transformer Vision transformer Supervised learning Transfer learning Self-supervised learning CNNs |
Language | English |
License | Copyright © 2023. Published by Elsevier B.V. |
LinkModel | OpenURL |
MergedId | FETCHMERGED-LOGICAL-c359t-5297b70feca097912b1973c13c10f4fb3894e07d780afb89b6d07fb6a139dc953 |
Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
ORCID | 0000-0002-3154-9851 0000-0002-7090-3825 0000-0001-5486-1613 |
PMID | 37924750 |
PQID | 2886328794 |
PQPubID | 23479 |
PageCount | 1 |
ParticipantIDs | proquest_miscellaneous_2886328794 crossref_primary_10_1016_j_media_2023_102988 pubmed_primary_37924750 |
PublicationCentury | 2000 |
PublicationDate | 2024-Jan 2024-01-00 20240101 |
PublicationDateYYYYMMDD | 2024-01-01 |
PublicationDate_xml | – month: 01 year: 2024 text: 2024-Jan |
PublicationDecade | 2020 |
PublicationPlace | Netherlands |
PublicationPlace_xml | – name: Netherlands |
PublicationTitle | Medical image analysis |
PublicationTitleAlternate | Med Image Anal |
PublicationYear | 2024 |
References | Serpen (10.1016/j.media.2023.102988_b58) 2008; 38 Hosseinzadeh Taher (10.1016/j.media.2023.102988_b33) 2021 Gutmann (10.1016/j.media.2023.102988_b27) 2010 Masutani (10.1016/j.media.2023.102988_b46) 2002; 21 10.1016/j.media.2023.102988_b12 U.S. Department of Health and Human Services Food and Drug Administration (10.1016/j.media.2023.102988_b72) 2008 Zhou (10.1016/j.media.2023.102988_b87) 2021; 67 Islam (10.1016/j.media.2023.102988_b39) 2021 Huang (10.1016/j.media.2023.102988_b36) 2020; 3 Zhuang (10.1016/j.media.2023.102988_b89) 2019 Asano (10.1016/j.media.2023.102988_b1) 2020 Caron (10.1016/j.media.2023.102988_b6) 2021 Han (10.1016/j.media.2023.102988_b30) 2021 Haghighi (10.1016/j.media.2023.102988_b29) 2021 Sun (10.1016/j.media.2023.102988_b64) 2017; 83 10.1016/j.media.2023.102988_b14 Tajbakhsh (10.1016/j.media.2023.102988_b65) 2015 Fairfield (10.1016/j.media.2023.102988_b19) 1990 Tajbakhsh (10.1016/j.media.2023.102988_b67) 2019; 58 Huang (10.1016/j.media.2023.102988_b37) 2020; 3 Jing (10.1016/j.media.2023.102988_b40) 2020 Stein (10.1016/j.media.2023.102988_b62) 2006; 354 Shi (10.1016/j.media.2023.102988_b61) 2020 Zbontar (10.1016/j.media.2023.102988_b81) 2021 10.1016/j.media.2023.102988_b21 Tajbakhsh (10.1016/j.media.2023.102988_b68) 2016; 35 Guo (10.1016/j.media.2023.102988_b26) 2022 Özkan (10.1016/j.media.2023.102988_b49) 2014; 113 Brown (10.1016/j.media.2023.102988_b4) 2020 Gidaris (10.1016/j.media.2023.102988_b22) 2018 Roth (10.1016/j.media.2023.102988_b55) 2014 Hu (10.1016/j.media.2023.102988_b34) 2021 Shamshad (10.1016/j.media.2023.102988_b60) 2022 Tajbakhsh (10.1016/j.media.2023.102988_b66) 2021; 40 González (10.1016/j.media.2023.102988_b24) 2020 Rajan (10.1016/j.media.2023.102988_b53) 2020 Scott (10.1016/j.media.2023.102988_b57) 1993; 186 Chen (10.1016/j.media.2023.102988_b10) 2020 Zhou (10.1016/j.media.2023.102988_b85) 2021 Wang (10.1016/j.media.2023.102988_b74) 2021 Cho (10.1016/j.media.2023.102988_b13) 2014 Gildenblat (10.1016/j.media.2023.102988_b23) 2021 Feng (10.1016/j.media.2023.102988_b20) 2020 Zbontar (10.1016/j.media.2023.102988_b82) 2021 Park (10.1016/j.media.2023.102988_b50) 2010; 58 10.1016/j.media.2023.102988_b77 Suman (10.1016/j.media.2023.102988_b63) 2021 10.1016/j.media.2023.102988_b31 10.1016/j.media.2023.102988_b75 Zhou (10.1016/j.media.2023.102988_b88) 2019 10.1016/j.media.2023.102988_b32 10.1016/j.media.2023.102988_b76 Lucassen (10.1016/j.media.2023.102988_b43) 2013; 131 Masoudi (10.1016/j.media.2023.102988_b45) 2018; 5 Tourassi (10.1016/j.media.2023.102988_b70) 1995; 194 Bi (10.1016/j.media.2023.102988_b3) 2007 Noroozi (10.1016/j.media.2023.102988_b48) 2016 Yang (10.1016/j.media.2023.102988_b78) 2019; 7 Roth (10.1016/j.media.2023.102988_b54) 2015; 35 Setio (10.1016/j.media.2023.102988_b59) 2016; 35 Zhou (10.1016/j.media.2023.102988_b84) 2009; 36 Chen (10.1016/j.media.2023.102988_b8) 2020 Liang (10.1016/j.media.2023.102988_b41) 2007 Ma (10.1016/j.media.2023.102988_b44) 2022 Ben-Cohen (10.1016/j.media.2023.102988_b2) 2016 10.1016/j.media.2023.102988_b38 Sanh (10.1016/j.media.2023.102988_b56) 2019 10.1016/j.media.2023.102988_b35 Litjens (10.1016/j.media.2023.102988_b42) 2017; 42 10.1016/j.media.2023.102988_b79 Touvron (10.1016/j.media.2023.102988_b71) 2021 Yu (10.1016/j.media.2023.102988_b80) 2021 Grill (10.1016/j.media.2023.102988_b25) 2020 Colak (10.1016/j.media.2023.102988_b15) 2021; 3 Deng (10.1016/j.media.2023.102988_b16) 2020 10.1016/j.media.2023.102988_b86 Zhou (10.1016/j.media.2023.102988_b83) 2021 Chen (10.1016/j.media.2023.102988_b11) 2020 Haghighi (10.1016/j.media.2023.102988_b28) 2020 Tian (10.1016/j.media.2023.102988_b69) 2020 Devlin (10.1016/j.media.2023.102988_b17) 2018 Dosovitskiy (10.1016/j.media.2023.102988_b18) 2020 10.1016/j.media.2023.102988_b5 10.1016/j.media.2023.102988_b9 10.1016/j.media.2023.102988_b7 10.1016/j.media.2023.102988_b47 Vaswani (10.1016/j.media.2023.102988_b73) 2017 Prasoon (10.1016/j.media.2023.102988_b52) 2013 Patil (10.1016/j.media.2023.102988_b51) 1993; 104 |
References_xml | – year: 2020 ident: 10.1016/j.media.2023.102988_b24 contributor: fullname: González – start-page: 85 year: 2020 ident: 10.1016/j.media.2023.102988_b20 article-title: Parts2Whole: Self-supervised contrastive learning via reconstruction contributor: fullname: Feng – year: 2008 ident: 10.1016/j.media.2023.102988_b72 contributor: fullname: U.S. Department of Health and Human Services Food and Drug Administration – ident: 10.1016/j.media.2023.102988_b35 doi: 10.1109/CVPR.2018.00745 – ident: 10.1016/j.media.2023.102988_b79 doi: 10.1109/CVPR.2017.75 – start-page: 712 year: 1990 ident: 10.1016/j.media.2023.102988_b19 article-title: Toboggan contrast enhancement for contrast segmentation contributor: fullname: Fairfield – start-page: 1597 year: 2020 ident: 10.1016/j.media.2023.102988_b10 article-title: A simple framework for contrastive learning of visual representations contributor: fullname: Chen – start-page: 246 year: 2013 ident: 10.1016/j.media.2023.102988_b52 article-title: Deep feature learning for knee cartilage segmentation using a triplanar convolutional neural network contributor: fullname: Prasoon – start-page: 3 year: 2021 ident: 10.1016/j.media.2023.102988_b33 article-title: A systematic benchmarking analysis of transfer learning for medical image analysis contributor: fullname: Hosseinzadeh Taher – year: 2021 ident: 10.1016/j.media.2023.102988_b83 contributor: fullname: Zhou – year: 2021 ident: 10.1016/j.media.2023.102988_b85 article-title: Active, continual fine tuning of convolutional neural networks for reducing annotation efforts publication-title: Med. Image Anal. contributor: fullname: Zhou – ident: 10.1016/j.media.2023.102988_b9 doi: 10.1109/CVPR46437.2021.01549 – ident: 10.1016/j.media.2023.102988_b31 doi: 10.1109/CVPR42600.2020.00975 – start-page: 1 year: 2020 ident: 10.1016/j.media.2023.102988_b40 article-title: Self-supervised visual feature learning with deep neural networks: A survey publication-title: TPAMI contributor: fullname: Jing – volume: 186 start-page: 661 issue: 3 year: 1993 ident: 10.1016/j.media.2023.102988_b57 article-title: Neural network analysis of ventilation-perfusion lung scans publication-title: Radiology doi: 10.1148/radiology.186.3.8430170 contributor: fullname: Scott – volume: 35 start-page: 1299 issue: 5 year: 2016 ident: 10.1016/j.media.2023.102988_b68 article-title: Convolutional neural networks for medical image analysis: Full training or fine tuning? publication-title: IEEE Trans. Med. Imaging doi: 10.1109/TMI.2016.2535302 contributor: fullname: Tajbakhsh – ident: 10.1016/j.media.2023.102988_b75 doi: 10.1109/CVPR.2018.00393 – ident: 10.1016/j.media.2023.102988_b86 doi: 10.1109/CVPR.2017.506 – volume: 3 start-page: 1 issue: 1 year: 2020 ident: 10.1016/j.media.2023.102988_b37 article-title: PENet—a scalable deep-learning model for automated diagnosis of pulmonary embolism using volumetric CT imaging publication-title: npj Digit. Med. contributor: fullname: Huang – ident: 10.1016/j.media.2023.102988_b38 doi: 10.1109/CVPR.2017.243 – volume: 83 start-page: 58 year: 2017 ident: 10.1016/j.media.2023.102988_b64 article-title: Automatic segmentation of liver tumors from multiphase contrast-enhanced CT images based on FCNs publication-title: Artif. Intell. Med. doi: 10.1016/j.artmed.2017.03.008 contributor: fullname: Sun – start-page: 420 year: 2019 ident: 10.1016/j.media.2023.102988_b89 article-title: Self-supervised feature learning for 3d medical images by playing a rubik’s cube contributor: fullname: Zhuang – start-page: 220 year: 2020 ident: 10.1016/j.media.2023.102988_b53 article-title: Pi-PE: A pipeline for pulmonary embolism detection using sparsely annotated 3D CT images contributor: fullname: Rajan – ident: 10.1016/j.media.2023.102988_b32 doi: 10.1109/CVPR.2016.90 – year: 2021 ident: 10.1016/j.media.2023.102988_b82 contributor: fullname: Zbontar – start-page: 743 year: 2020 ident: 10.1016/j.media.2023.102988_b61 article-title: Automatic diagnosis of pulmonary embolism using an attention-guided framework: A large-scale study contributor: fullname: Shi – start-page: 5998 year: 2017 ident: 10.1016/j.media.2023.102988_b73 article-title: Attention is all you need contributor: fullname: Vaswani – year: 2020 ident: 10.1016/j.media.2023.102988_b4 contributor: fullname: Brown – year: 2022 ident: 10.1016/j.media.2023.102988_b60 contributor: fullname: Shamshad – volume: 36 start-page: 3385 issue: 8 year: 2009 ident: 10.1016/j.media.2023.102988_b84 article-title: Computer-aided detection of pulmonary embolism in computed tomographic pulmonary angiography (CTPA): Performance evaluation with independent data sets publication-title: Med. Phys. doi: 10.1118/1.3157102 contributor: fullname: Zhou – ident: 10.1016/j.media.2023.102988_b5 doi: 10.1007/978-3-030-01264-9_9 – ident: 10.1016/j.media.2023.102988_b47 doi: 10.1109/CVPR42600.2020.00674 – year: 2020 ident: 10.1016/j.media.2023.102988_b11 contributor: fullname: Chen – start-page: 66 year: 2022 ident: 10.1016/j.media.2023.102988_b26 article-title: Discriminative, restorative, and adversarial learning: Stepwise incremental pretraining contributor: fullname: Guo – year: 2020 ident: 10.1016/j.media.2023.102988_b1 contributor: fullname: Asano – volume: 35 start-page: 1170 issue: 5 year: 2015 ident: 10.1016/j.media.2023.102988_b54 article-title: Improving computer-aided detection using convolutional neural networks and random view aggregation publication-title: IEEE Trans. Med. Imaging doi: 10.1109/TMI.2015.2482920 contributor: fullname: Roth – ident: 10.1016/j.media.2023.102988_b7 doi: 10.1109/ICCV48922.2021.00951 – start-page: 520 year: 2014 ident: 10.1016/j.media.2023.102988_b55 article-title: A new 2.5 D representation for lymph node detection using random sets of deep convolutional neural network observations contributor: fullname: Roth – volume: 58 year: 2019 ident: 10.1016/j.media.2023.102988_b67 article-title: Computer-aided detection and visualization of pulmonary embolism using a novel, compact, and discriminative image representation publication-title: Med. Image Anal. doi: 10.1016/j.media.2019.101541 contributor: fullname: Tajbakhsh – start-page: 384 year: 2019 ident: 10.1016/j.media.2023.102988_b88 article-title: Models genesis: Generic autodidactic models for 3D medical image analysis contributor: fullname: Zhou – ident: 10.1016/j.media.2023.102988_b12 doi: 10.1109/ICCV48922.2021.00950 – volume: 67 year: 2021 ident: 10.1016/j.media.2023.102988_b87 article-title: Models genesis publication-title: Med. Image Anal. doi: 10.1016/j.media.2020.101840 contributor: fullname: Zhou – start-page: 137 year: 2020 ident: 10.1016/j.media.2023.102988_b28 article-title: Learning semantics-enriched representation via self-discovery, self-classification, and self-restoration contributor: fullname: Haghighi – start-page: 62 year: 2015 ident: 10.1016/j.media.2023.102988_b65 article-title: Computer-aided pulmonary embolism detection using a novel vessel-aligned multi-planar image representation and convolutional neural networks contributor: fullname: Tajbakhsh – year: 2021 ident: 10.1016/j.media.2023.102988_b74 contributor: fullname: Wang – year: 2020 ident: 10.1016/j.media.2023.102988_b18 contributor: fullname: Dosovitskiy – volume: 58 start-page: 1519 issue: 6 year: 2010 ident: 10.1016/j.media.2023.102988_b50 article-title: A multistage approach to improve performance of computer-aided detection of pulmonary embolisms depicted on CT images: Preliminary investigation publication-title: IEEE Trans. Biomed. Eng. doi: 10.1109/TBME.2010.2063702 contributor: fullname: Park – start-page: 1 year: 2020 ident: 10.1016/j.media.2023.102988_b16 article-title: Deep learning in digital pathology image analysis: A survey publication-title: Front. Med. contributor: fullname: Deng – start-page: 69 year: 2016 ident: 10.1016/j.media.2023.102988_b48 article-title: Unsupervised learning of visual representations by solving jigsaw puzzles contributor: fullname: Noroozi – year: 2020 ident: 10.1016/j.media.2023.102988_b8 contributor: fullname: Chen – volume: 38 start-page: 204 issue: 2 year: 2008 ident: 10.1016/j.media.2023.102988_b58 article-title: A knowledge-based artificial neural network classifier for pulmonary embolism diagnosis publication-title: Comput. Biol. Med. doi: 10.1016/j.compbiomed.2007.10.001 contributor: fullname: Serpen – year: 2020 ident: 10.1016/j.media.2023.102988_b25 contributor: fullname: Grill – volume: 7 start-page: 84849 year: 2019 ident: 10.1016/j.media.2023.102988_b78 article-title: A two-stage convolutional neural network for pulmonary embolism detection from CTPA images publication-title: IEEE Access doi: 10.1109/ACCESS.2019.2925210 contributor: fullname: Yang – start-page: 692 year: 2021 ident: 10.1016/j.media.2023.102988_b39 article-title: Seeking an optimal approach for computer-aided pulmonary embolism detection contributor: fullname: Islam – year: 2021 ident: 10.1016/j.media.2023.102988_b23 contributor: fullname: Gildenblat – year: 2021 ident: 10.1016/j.media.2023.102988_b6 contributor: fullname: Caron – volume: 40 start-page: 2526 issue: 10 year: 2021 ident: 10.1016/j.media.2023.102988_b66 article-title: Guest editorial annotation-efficient deep learning: The holy grail of medical imaging publication-title: IEEE Trans. Med. Imaging doi: 10.1109/TMI.2021.3089292 contributor: fullname: Tajbakhsh – year: 2021 ident: 10.1016/j.media.2023.102988_b34 contributor: fullname: Hu – year: 2018 ident: 10.1016/j.media.2023.102988_b17 contributor: fullname: Devlin – volume: 3 issue: 2 year: 2021 ident: 10.1016/j.media.2023.102988_b15 article-title: The RSNA pulmonary embolism CT dataset publication-title: Radiol. Artif. Intell. doi: 10.1148/ryai.2021200254 contributor: fullname: Colak – start-page: 297 year: 2010 ident: 10.1016/j.media.2023.102988_b27 article-title: Noise-contrastive estimation: A new estimation principle for unnormalized statistical models contributor: fullname: Gutmann – volume: 131 start-page: 145 issue: 2 year: 2013 ident: 10.1016/j.media.2023.102988_b43 article-title: Concerns in using multi-detector computed tomography for diagnosing pulmonary embolism in daily practice. A cross-sectional analysis using expert opinion as reference standard publication-title: Thromb. Res. doi: 10.1016/j.thromres.2012.11.027 contributor: fullname: Lucassen – volume: 194 start-page: 889 issue: 3 year: 1995 ident: 10.1016/j.media.2023.102988_b70 article-title: Artificial neural network for diagnosis of acute pulmonary embolism: effect of case and observer selection publication-title: Radiology doi: 10.1148/radiology.194.3.7862997 contributor: fullname: Tourassi – volume: 3 start-page: 1 issue: 1 year: 2020 ident: 10.1016/j.media.2023.102988_b36 article-title: PENet—a scalable deep-learning model for automated diagnosis of pulmonary embolism using volumetric CT imaging publication-title: npj Digit. Med. contributor: fullname: Huang – ident: 10.1016/j.media.2023.102988_b14 doi: 10.1109/CVPR.2017.195 – ident: 10.1016/j.media.2023.102988_b76 doi: 10.1109/WACV56688.2023.00358 – start-page: 1 year: 2007 ident: 10.1016/j.media.2023.102988_b3 article-title: Multiple instance learning of pulmonary embolism detection with geodesic distance along vascular structure contributor: fullname: Bi – year: 2014 ident: 10.1016/j.media.2023.102988_b13 contributor: fullname: Cho – ident: 10.1016/j.media.2023.102988_b21 doi: 10.1109/CVPR46437.2021.00676 – start-page: 356 year: 2021 ident: 10.1016/j.media.2023.102988_b63 article-title: Attention based CNN-LSTM network for pulmonary embolism prediction on chest computed tomography pulmonary angiograms contributor: fullname: Suman – volume: 5 start-page: 1 issue: 1 year: 2018 ident: 10.1016/j.media.2023.102988_b45 article-title: A new dataset of computed-tomography angiography images for computer-aided detection of pulmonary embolism publication-title: Sci. Data doi: 10.1038/sdata.2018.180 contributor: fullname: Masoudi – volume: 21 start-page: 1517 issue: 12 year: 2002 ident: 10.1016/j.media.2023.102988_b46 article-title: Computerized detection of pulmonary embolism in spiral CT angiography based on volumetric image analysis publication-title: IEEE TMI contributor: fullname: Masutani – start-page: 10347 year: 2021 ident: 10.1016/j.media.2023.102988_b71 article-title: Training data-efficient image transformers & distillation through attention contributor: fullname: Touvron – start-page: 45 year: 2021 ident: 10.1016/j.media.2023.102988_b80 article-title: Mil-vt: Multiple instance learning enhanced vision transformer for fundus image classification contributor: fullname: Yu – volume: 42 start-page: 60 year: 2017 ident: 10.1016/j.media.2023.102988_b42 article-title: A survey on deep learning in medical image analysis publication-title: Med. Image Anal. doi: 10.1016/j.media.2017.07.005 contributor: fullname: Litjens – ident: 10.1016/j.media.2023.102988_b77 doi: 10.1109/CVPR.2017.634 – start-page: 12 year: 2022 ident: 10.1016/j.media.2023.102988_b44 article-title: Benchmarking and boosting transformers for medical image classification contributor: fullname: Ma – start-page: 77 year: 2016 ident: 10.1016/j.media.2023.102988_b2 article-title: Fully convolutional network for liver segmentation and lesions detection contributor: fullname: Ben-Cohen – volume: 354 start-page: 2317 issue: 22 year: 2006 ident: 10.1016/j.media.2023.102988_b62 article-title: Multidetector computed tomography for acute pulmonary embolism publication-title: N. Engl. J. Med. doi: 10.1056/NEJMoa052367 contributor: fullname: Stein – volume: 113 start-page: 757 issue: 3 year: 2014 ident: 10.1016/j.media.2023.102988_b49 article-title: A novel method for pulmonary embolism detection in CTA images publication-title: Comput. Methods Programs Biomed. doi: 10.1016/j.cmpb.2013.12.014 contributor: fullname: Özkan – volume: 104 start-page: 1685 issue: 6 year: 1993 ident: 10.1016/j.media.2023.102988_b51 article-title: Neural network in the clinical diagnosis of acute pulmonary embolism publication-title: Chest doi: 10.1378/chest.104.6.1685 contributor: fullname: Patil – year: 2019 ident: 10.1016/j.media.2023.102988_b56 contributor: fullname: Sanh – year: 2018 ident: 10.1016/j.media.2023.102988_b22 contributor: fullname: Gidaris – year: 2021 ident: 10.1016/j.media.2023.102988_b30 contributor: fullname: Han – start-page: 630 year: 2007 ident: 10.1016/j.media.2023.102988_b41 article-title: Computer aided detection of pulmonary embolism with tobogganing and multiple instance classification in CT pulmonary angiography contributor: fullname: Liang – year: 2021 ident: 10.1016/j.media.2023.102988_b29 contributor: fullname: Haghighi – year: 2020 ident: 10.1016/j.media.2023.102988_b69 contributor: fullname: Tian – year: 2021 ident: 10.1016/j.media.2023.102988_b81 contributor: fullname: Zbontar – volume: 35 start-page: 1160 issue: 5 year: 2016 ident: 10.1016/j.media.2023.102988_b59 article-title: Pulmonary nodule detection in CT images: false positive reduction using multi-view convolutional networks publication-title: IEEE Trans. Med. Imaging doi: 10.1109/TMI.2016.2536809 contributor: fullname: Setio |
SSID | ssj0007440 |
Score | 2.4814231 |
Snippet | Pulmonary Embolism (PE) represents a thrombus ("blood clot"), usually originating from a lower extremity vein, that travels to the blood vessels in the lung,... |
SourceID | proquest crossref pubmed |
SourceType | Aggregation Database Index Database |
StartPage | 102988 |
SubjectTerms | Computers Diagnosis, Computer-Assisted - methods Humans Imaging, Three-Dimensional Neural Networks, Computer Pulmonary Embolism - diagnostic imaging |
Title | Seeking an optimal approach for Computer-aided Diagnosis of Pulmonary Embolism |
URI | https://www.ncbi.nlm.nih.gov/pubmed/37924750 https://search.proquest.com/docview/2886328794 |
Volume | 91 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV3db9MwELegkxA8IBhf40tG4q0kSmzHjh83NJgQm5C2oomXyE7staNLEKRM8NdztpO0ZQMxpCqqrOhi-X653Nl3v0PopSCUa2FNpA2hEbNGRopUyvU3IVyBw5F4KqX9A743Ye-Os-NlCYGvLml1XP68tK7kf7QKY6BXVyV7Bc0OQmEA_oN-4Qoahus_6fjQmM-hyHDcwKt_5ur-O45wnz7Yt2yIHA9kBebNp9UFBpIPiznM0-XM7Z7pZt4zCfbNnbrzG5B54g4YAnPJRRyp6awaT4YsjU_TZuGPO5r65NzMhuQeM5033tofTmffh9GmPVc_VnL3xzvx6iYEYSubEMFuUp5GOQuVmb1hDW24OsuYOq73_FKjHfYPTmNfKxO7fu7x8u51iuzfPl1DQmGfq3ZaeCGFE1IEIdfRBhEyy0Zo4-3O-4_bw3faUSOGqrww9Z6Tymf_XZjLut_yh2DEOyVHd9DtLprA2wEad9E1U2-iWysck5voxn6XPXEPHXR4warGHV5wjxcMeMHreMEDXnBj8YAX3OPlPpq82T16vRd1_TSikmayjTIihRaJNaVKpJAp0akUtEzhl1hmNfiuzCSiEnmirM6l5lUirOYKooSqlBl9gEZ1U5tHCEOQrXJuK86ZZTKtckap4jaXJU11lrAt9KpfreJLoE0p_qKjLfSiX9ECzJs7s1K1aRbfCpLnnEJUL0Hkw7DUg0AqJGHg8T6-2sOeoJtLCD9Fo_brwjwDz7LVzzuI_AIzCnc2 |
link.rule.ids | 315,786,790,27955,27956 |
linkProvider | Elsevier |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Seeking+an+optimal+approach+for+Computer-aided+Diagnosis+of+Pulmonary+Embolism&rft.jtitle=Medical+image+analysis&rft.au=Islam%2C+Nahid+Ul&rft.au=Zhou%2C+Zongwei&rft.au=Gehlot%2C+Shiv&rft.au=Gotway%2C+Michael+B.&rft.date=2024-01-01&rft.issn=1361-8415&rft.volume=91&rft.spage=102988&rft_id=info:doi/10.1016%2Fj.media.2023.102988&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_media_2023_102988 |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1361-8415&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1361-8415&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1361-8415&client=summon |